A Tumor-Infiltrating Lymphocyte-Based Risk Model Predicts Prognosis of Colon Cancer

Tumor-infiltrating lymphocytes are relevant to the tumor prognosis and response to immunotherapy in colon cancer. The gene expression data of colon cancer was obtained from the cancer genome atlas (TCGA) database and the components of immune cell types were analyzed by CIBERSORT. Selection operator (LASSO) and multivariate Cox regression filtered immune cells and selected the most significant cell types to construct an immune risk model including memory B cells, plasma cells, T follicular helper (Tfh) cells, M0 macrophages and resting dendritic cells. Receiver operating characteristic (ROC) curves were used to verify the sensitivity and specificity of the model, which was validated in Gene Expression Omnibus (GEO) dataset. Combined with the clinical traits, a nomogram was established to predict the prognosis of colon cancer. According to function analyses through weighted correlation network analysis (WGCNA) and gene set enrichment analysis (GSEA), tumor-infiltrating immune cells showed significant importance in tumor immune-associated regulation, especially the adhesion, migration and invasion in colon cancer.